List your company

Interactive Graph Navigation on IGN Qwen3.5-8B backbone (evaluation)

39SR@1.5 (Easy)TANDEM
16.6422.44528.2534.055Jun 2, 2026

Evaluation results

MethodLinks
TANDEMBackbone=Qwen3.5-8B, Protocol=Cost-Aware QA, spmem=true, QA=true2026.06392.3235.83.13315.3314.222.335.35066.572.52.6823.870.220.971.281.13.5721.440.8
TANDEM w/o spmemBackbone=Qwen3.5-8B, Protocol=Cost-Aware QA, spmem=false, QA=true2026.0631.22.1292.49244.568.8162843.557.8653.1524.080.2610.820.933.0118.536
GTABackbone=Qwen3.5-8B, Protocol=Naive QA (unconstrained)2026.0629.76.9327.68.1722.49.768.817.626.739.350.657.93.4927.621.731.462.872.198.256.412.2
GTABackbone=Qwen3.5-8B, Protocol=Cost-Aware QA2026.0629.22.24273.0121.85.1410.517.22638.649.356.13.5526.850.240.951.21.053.4415.930.3
TANDEM w/o QABackbone=Qwen3.5-8B, Protocol=Cost-Aware QA, spmem=true, QA=false2026.0624.5—17—17.8—3.210.5203449.258.43.3517.27—————14.234.2
COINBackbone=Qwen3.5-8B, Protocol=Naive QA (unconstrained)2026.0623.67.4818.79.0314.310.815.912.418.629.236.846.34.6325.182.041.173.582.279.063.77.3
MapGPTBackbone=Qwen3.5-8B, Protocol=Naive QA (unconstrained)2026.0620.17.2717.18.9112.710.466.712.115.928.332.641.74.8929.071.841.393.112.538.873.67.1
MapGPTBackbone=Qwen3.5-8B, Protocol=Cost-Aware QA2026.0619.62.1517.42.8712.24.876.612.316.527.631.540.24.9728.680.260.931.10.993.2810.219.4
NavGPTBackbone=Qwen3.5-8B, Protocol=Naive QA (unconstrained)2026.0618.17.3415.68.9910.910.685.111.414.824.628.936.75.2331.641.991.183.242.619.023.16.1
COINBackbone=Qwen3.5-8B, Protocol=Cost-Aware QA2026.06181.3712.52.198.54.3323.5132227.435.95.228.40.45101.162.619.219.3
TANDEM w/o spmem & QABackbone=Qwen3.5-8B, Protocol=Cost-Aware QA, spmem=false, QA=false2026.0618—14—10—1.56142534.244.15.1521.16—————1023.8
NavGPTBackbone=Qwen3.5-8B, Protocol=Cost-Aware QA2026.0617.52.1115.32.8210.14.765.810.814.324.127.835.35.3129.150.270.921.060.963.218.916.9